aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim/Region/Physics/OdePlugin (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Move update of the final optional ODE total frame stat inside the OdeLock ↵Justin Clark-Casey (justincc)2012-06-291-3/+3
| | | | rather than outside to avoid a very occasional race condition with the stat collection thread
* refactor: rename _collisionEventPrim to m_collisionEventActors and ↵Justin Clark-Casey (justincc)2012-06-291-12/+12
| | | | | | _collisionEventPrimChanges to m_collisionEventActorsChanges to reflect their actual contents. These dictionaries handle all actor types, not just physical prims.
* Fix a regression where we stopped removing avatars from collision event ↵Justin Clark-Casey (justincc)2012-06-291-1/+1
| | | | | | | reporting on logout, rather than stopping clearing their collision events. This occurred in b18c8c8 (Thu May 17 2012). This was a cause of very occasional race conditions and likely memory leakage as clients came and went from the region.
* Add optional total avatars, total prims and active prims stats to ODE plugin.Justin Clark-Casey (justincc)2012-06-011-0/+21
| | | | | These will act as a sanity check with the main scene stats, to show that physics scene entities are being managed properly. Total prims will not match scene total prims since physics total does not include phantom prims
* If OdeScene.Near() returns no collision contacts, then exit as early as ↵Justin Clark-Casey (justincc)2012-06-011-0/+4
| | | | possible. All subsequent code is only relevant if there are contacts.
* Add optional stat for the other collision time per frame not spent in ODE ↵Justin Clark-Casey (justincc)2012-06-011-1/+16
| | | | native spaces or geom collision code
* Add avatar forces calculation, prim force and raycasting per frame ↵Justin Clark-Casey (justincc)2012-06-011-7/+46
| | | | millisecond optional stats
* Collection optional avatar and prim taint frame millisecond timesJustin Clark-Casey (justincc)2012-06-011-7/+36
|
* Remove recent optional native collision frame milliseconds statJustin Clark-Casey (justincc)2012-06-011-13/+1
| | | | Unnecessary since this has now been broken down into space collisions and geom collisions
* Add optional stat that records milliseconds spent notifying collision ↵Justin Clark-Casey (justincc)2012-06-011-0/+14
| | | | listeners in physics frames
* Add avatar and prim update milliseconds per frame optional statsJustin Clark-Casey (justincc)2012-06-011-7/+31
|
* Add option native step frame ms statJustin Clark-Casey (justincc)2012-06-011-9/+21
|
* Add total ODE frame time optional stat, as a sanity check on the main scene ↵Justin Clark-Casey (justincc)2012-06-011-7/+21
| | | | physics stat
* Break down native ODE collision frame time stat into native space collision ↵Justin Clark-Casey (justincc)2012-06-011-4/+22
| | | | and geom collision stats
* Rename new collision stats to 'contacts' - there are/can be multiple ↵Justin Clark-Casey (justincc)2012-06-011-6/+6
| | | | contacts per collision and this is what is actually being measured.
* Stop adding an unnecessary duplicate _perloopcontact if the avatar is ↵Justin Clark-Casey (justincc)2012-06-011-9/+4
| | | | | | standing on a prim. This has already been added earlier on in the method.
* minor: comment out currently unused OdeScene.sCollisionDataJustin Clark-Casey (justincc)2012-06-011-9/+9
|
* Add ODE avatar and prim collision numbers if extra stats collection is enabled.Justin Clark-Casey (justincc)2012-06-011-5/+34
|
* Fix OdeScene.GetTopColliders() to return the top 25 colliders rather than ↵Justin Clark-Casey (justincc)2012-06-011-19/+13
| | | | | | | | the first 25 that had non-zero collision scores. Also zeros collisions scores on all prims after report collection, not just the top 25. As before, this collision scores are only reset after a report is requested, which may give unrealistic numbers on the first request. So to see more realistic scores, ignore the first report and then refresh the request after a couple of seconds or so.
* Add an optional mechanism for physics modules to collect and return ↵Justin Clark-Casey (justincc)2012-05-311-10/+129
| | | | | | | | | | | arbitrary stats. If active, the physics module can return arbitrary stat counters that can be seen via the MonitoringModule (http://opensimulator.org/wiki/Monitoring_Module) This is only active in OdeScene if collect_stats = true in [ODEPhysicsSettings]. This patch allows OdeScene to collect elapsed time information for calls to the ODE native collision methods to assess what proportion of time this takes compared to total physics processing. This data is returned as ODENativeCollisionFrameMS in the monitoring module, updated every 3 seconds. The performance effect of collecting stats is probably extremely minor, dwarfed by the rest of the physics code.
* refactor: move EventQueueGet path generation into common method. Rename ↵Justin Clark-Casey (justincc)2012-05-181-1/+1
| | | | some local variables in line with code conventions. Add commented out EQG log lines for future use.
* Don't eagerly clear frame collision events when physics actors subscribe and ↵Justin Clark-Casey (justincc)2012-05-171-2/+8
| | | | | | | | unsubscribe from collisions, in order to avoid a race condition. Since this is done directly from ScenePresence, it can lead to a race condition with the simulator loop. There's no real point doing it anyway since the clear will be done very shortly afterwards by the simulate loop and either there are no events (for a new avatar) or events don't matter (for a departing avatar). This matches existing behaviour in OdePrim
* Restore _parent_scene.actor_name_map[prim_geom] = this; accidentally removed ↵Justin Clark-Casey (justincc)2012-04-241-0/+1
| | | | | | | | from ODEPrim.SetGeom. This occurred in 7a574be3fd from Sat 21 Apr 2012. This should fix collision detection. Mnay thanks to tglion for the spot and the fix in http://opensimulator.org/mantis/view.php?id=5988
* Comment out spurious Body != IntPtr.Zero code after disableBody(), since ↵Justin Clark-Casey (justincc)2012-04-212-10/+20
| | | | disableBody() sets Body == IntPtr.Zero on all code paths.
* refactor: Simplify ODEPrim.AddChildPrim() by returning early where appropriate.Justin Clark-Casey (justincc)2012-04-211-115/+117
|
* Remove redundant prim_geom != IntPtr.Zero checks in ODEPrim.Justin Clark-Casey (justincc)2012-04-211-116/+88
| | | | | | prim_geom == IntPtr.Zero only before a new add prim taint is processed (which is the first taint) or in operations such as scale change which are done in taint or under lock. Therefore, we can remove these checks which were not consistently applied anyway. If there is a genuine problem, better to see it quickly in a NullReferenceException than hide the bug.
* If a physical prim is manually moved (e.g. by a user) then set the geometry ↵Justin Clark-Casey (justincc)2012-04-201-12/+10
| | | | | | | position as well as the body position This is necessary to stop the moved prim snapping back to the original position on deselection if moved only once This resolves http://opensimulator.org/mantis/view.php?id=5966
* refactor: precalculate the fixed movement factor for avatar tilting ↵Justin Clark-Casey (justincc)2012-03-202-8/+16
| | | | (sqrt(2)) rather than doing it multiple times on every move.
* refactor: Eliminate unnecessary duplicate avCapsuleTiltedJustin Clark-Casey (justincc)2012-03-201-3/+7
|
* Fix crash where two scene loop threads could changes m_MeshToTriMeshMap at ↵Justin Clark-Casey (justincc)2012-03-201-10/+16
| | | | | | the same time. Have to lock m_MeshToTriMeshMap as property is static and with more than one region two scene loops could try to manipulate at the same time.
* Stop spurious scene loop startup timeout alarms for scenes with many prims.Justin Clark-Casey (justincc)2012-02-241-0/+2
| | | | | | | | On the first frame, all startup scene objects are added to the physics scene. This can cause a considerable delay, so we don't start raising the alarm on scene loop timeouts until the second frame. This commit also slightly changes the behaviour of timeout reporting. Previously, a report was made for the very first timed out thread, ignoring all others until the next watchdog check. Instead, we now report every timed out thread, though we still only do this once no matter how long the timeout.
* Add 64 bit Windows sqlite3.dll and use this if running a 64-bit windows process.Justin Clark-Casey (justincc)2012-02-211-14/+1
|
* Report an error if the required native Windows ODE library can't be found.Justin Clark-Casey (justincc)2012-02-181-1/+4
|
* On Windows automatically load the correct native ODE library depending on ↵Justin Clark-Casey (justincc)2012-02-181-2/+22
| | | | | | | | | whether the process is 32-bit or 64-bit In theory, this means that a 64-bit Windows OS user can now run OpenSim.exe with ODE and use more than 2 (or 3) GB of memory. However, this is completely untested since I don't currently own a 64-bit Windows box. Feedback appreciated. Using OpenSim.32BitLaunch.exe should continue to work. Other platforms are unaffected. This will currently not work with sqlite - I will add that too if this works.
* correct the default avatar_terminal_velocity value that I accidentally left ↵Justin Clark-Casey (justincc)2012-02-111-1/+1
| | | | in whilst testing
* Clamp ODE character velocity. Make ODE falling character 54m/s by default.Justin Clark-Casey (justincc)2012-02-112-1/+53
| | | | | | | If velocity reaches 256 in any vector then bad things happen with ODE, so we now clamp this value. In addition, a falling avatar is clamped by default at 54 m/s, which is the same as a falling skydiver. This also appears to be the value used on the linden lab grid. This should resolve http://opensimulator.org/mantis/view.php?id=5882
* Serialize calls to ODE Collide() function across OdeScene instances to ↵Justin Clark-Casey (justincc)2012-01-252-1/+32
| | | | | | | | | prevent ODE crashes on simulators running more than one region. It turns out that calls to Collide() are not thread-safe even for objects in different ODE physics worlds due to ODE static caches. For simulators running multiple regions, not serializing calls from different scene loops will sooner or later cause OpenSim to crash with a native stack trace referencing OBBCollider. This affects the default OPCODE collider but not GIMPACT. However, GIMPACT fails for other reasons under some current simulator loads. ODE provides a thread local storage option, but as of ODE r1755 (and r1840) DLLs compiled with this crash OpenSim immediately.
* Restrict accessible of ODECharacter Shell and Body. Add method doc and some ↵Justin Clark-Casey (justincc)2012-01-242-12/+27
| | | | error log lines.
* Adding commented out log messages and some minor formatting for future bug ↵Justin Clark-Casey (justincc)2012-01-021-0/+1
| | | | hunting. No functional changes.
* Align default ODE_STEPSIZE with that already used through OpenSimDefaults.iniJustin Clark-Casey (justincc)2011-12-221-2/+2
|
* Remove unused SetAcceleration and add set on Acceleration parameterDan Lake2011-12-202-6/+2
|
* Merge branch 'master' of ssh://opensimulator.org/var/git/opensimDiva Canto2011-12-201-1/+1
|\
| * Comment out accidentally left in "Adding physics prim" log messageJustin Clark-Casey (justincc)2011-12-161-1/+1
| |
* | Fixed bug of avie going under the terrain when crossing regions in certain ↵Diva Canto2011-12-201-2/+2
|/ | | | directions. This was a 1-off bug: the terrain was being placed in 127, 127 resulting in a bounding box if -2, 256. I placed it in 128, 128 resulting in a bounding box of -1, 257.
* Stop pointlessly setting the m_colliderarr[] to false in the ODECharacter ↵Justin Clark-Casey (justincc)2011-12-151-5/+0
| | | | constructor
* Stop having to call SetHeight again in ScenePresence.AddToPhysicalScene() ↵Justin Clark-Casey (justincc)2011-12-151-16/+22
| | | | | | when we've already passed size information to the avatar at PhysicsScene.AddAvatar() Eliminate some copypasta for height setting in OdeCharacter
* Code cleanup related to ScenePresence.PhysicsActor and OdeScene/OdeCharacterJustin Clark-Casey (justincc)2011-12-152-8/+24
| | | | | | Stop hiding RemoveAvatar failure, add log messages when characters are removed through defects or re-added unexpectedly. Add commented out log lines for future use. Use automatic property for PhysicsActor for better code readability and simplicity
* minor: add some currently commented log lines for use in debuggingJustin Clark-Casey (justincc)2011-12-141-1/+8
|
* Simplify some manipulation of _taintedActors in OdeSceneJustin Clark-Casey (justincc)2011-12-141-13/+6
|
* Eliminate _taintedPrimsH and _taintedPrimsL (and _taintedPrimLock) in favour ↵Justin Clark-Casey (justincc)2011-12-141-45/+14
| | | | | | of just a _taintedPrims HashSet. There's no point maintaining a list because any pending taint operations are all carried out in the same call anyway.